Network performance monitoring systems often use distributed active “probes” that inject traffic into the network. Probes may be either dedicated hardware devices or software running on devices or entities in the network, such as personal computers, telephones and gateways. The software running on a device functioning as a probe to generate test traffic is known as a “test plug.”
Using the entire population of possible probes is usually unnecessary, as during a typical network performance test, statistical information is being collected. Furthermore, using the entire population of possible probes has the potential of overloading the network. Accordingly, network performance monitoring systems often need to select a subset of all possible probes as the active set.
The selection of probes has generally been done through manual configuration or on a first-come, first-served basis. However, using such techniques, it is difficult or impossible to ensure that the set of active probes is representative of the total population of devices or entities on the network. In particular, characteristics of devices that may not be represented or that may not be represented in a realistic and/or desired proportion to all devices on the network include device type, network (or subnetwork) location and traffic parameters.
More particularly, techniques for use in connection with network performance monitoring have used the Internet protocol (IP) address to identify probes. Furthermore, simple network management protocol (SNMP) has been used to collect data from the probes. However, in order to monitor a converged network, where the network devices have more than one identity, the IP address by itself is not sufficient to identify the network probes. In particular, in a converged network, probes may have both an IP address for use in connection with the data network domain and an extension number or multimedia identity in a telephony or multimedia domain.